Drug-discrimination procedures with animals have been used to model drug-induced changes in mood in humans. Our previous research, using a two-choice discrimination, has shown that time-dependent, bidirectional changes in choice responding occur following administration of drug. We have also shown that knowledge of the time course of this bidirectional shift is important for understanding the effects of repeated drug administration as well as agonist-antagonist interactions. These findings suggest that the phenomena of pharmacodynamic tolerance, withdrawal, and drug-dependence can be understood in terms of normal, adaptive processes. Research is needed, however, to establish the degree of specificity of the rebound or adaptive state that follows the primary drug effect. Because both the drug-induced primary cue state and any subsequent rebound state must be defined relative to the "normal" state, unambiguous definition of "normal" is required. The proposed research will do this by using a three-choice drug- discrimination task that includes a specific saline-appropriate response alternative. The initial experiment will characterize the time- dependent changes in cue state following single doses of amphetamine and cocaine. Then, the extent to which these bidirectional changes mediate drug withdrawal and drug tolerance will be determined. The results of these studies will define parameters that will then be used to assess the extent to which interdose intervals can be chosen to either optimize or preclude tolerance resulting from repeated drug administration. The final experiment will determine if the post-amphetamine withdrawal state is an effective cue for discrimination learning. If learning occurs as expected, the withdrawal state will be further characterized by identifying drugs with cue properties that will substitute for it. The unique aspects of this research are its ability to measure adaptive processes that oppose the primary effect of psychoactive drugs, and its emphasis on characterizing the effects of drugs in terms of bidirectional changes over time. The results of the studies proposed should (1) contribute to an understanding of adaptive processes underlying tolerance, withdrawal, and sensitization, and (2) help establish the relevance of the drug-discrimination model to questions of drug use by humans.